factors predicitve of subjective sleep quality and of

FACTORS PREDICITVE OF SUBJECTIVE SLEEP QUALITY AND OF INSOMNIA
COMPLAINT IN PATIENTS WITH STROKE: IMPLICATIONS FOR CLINICAL
PRACTICE
FATORES PREDITIVOS DA QUALIDADE SUBJETIVA DO SONO E DAS
QUEIXAS DE INSÔNIA EM PACIENTES COM ACIDENTE VASCULAR
ENCEFÁLICO: IMPLICAÇÕES PARA A PRÁTICA CLÍNICA
TITLE RUNNIG: SLEEP QUALITY AND INSOMNIA IN THE STROKE
TÍTULO RESUMIDO: QUALIDADE DO SONO E INSÔNIA NO AVC
Patrícia Cavalcanti da Rocha, Marina Tostes Miranda Barroso, Ana Amália Torres Souza
Gandour Dantas, Luciana Protásio de Melo, Tania Fernandes Campos.
Universidade Federal do Rio Grande do Norte
Seção da Academia na qual se enquadra o artigo: Ciências Biomédicas e Médicas
Autor correspondência: Tania Fernandes Campos. Universidade Federal do Rio Grande do
Norte, Departamento de Fisioterapia, Avenida Senador Salgado Filho, 3000, CEP: 59066800, Natal/RN, Brasil. Telefax: 84-3342-2001. E-mail: [email protected]
1
ABSTRACT
The complaints regarding sleep problems have not been well identified after a stroke. The aim
of this study was to investigate the predictive factors of sleep quality and of insomnia
complaints in patients with stroke. A total of 70 subjects, 40 patients (57 ± 7 years) and 30
healthy controls (52 ± 6 years), assessed by the Pittsburg Sleep Quality Index (PSQI) and the
Sleep Habits Questionnaire, took part in the study. The data were analyzed using the chisquare test, the Student’s t-test and logistic regression analysis. On average, the patients
showed poor sleep quality (patients: 6.3 ± 3.5; controls: 3.9 ± 2.2; p= 0.002) and insomnia
complaint was the most prevalent (patients: 37.5%; controls: 6.7%; p= 0.007). The absence of
insomnia complaint (OR= 0.120; 95%CI= 0.017-0.873; p= 0.036) and the decreased latency
of sleep (OR= 0.120; 95%CI= 0.017-0.873; p= 0.036) were the protective factors of sleep
quality. Female sex (OR= 11.098; 95%CI= 1.167-105.559; p= 0.036) and fragmented sleep
(OR= 32.040; 95%CI= 3.236-317.261; p= 0.003) were the risk factors for insomnia
complaint. We suggest that complaints of poor sleep quality and of insomnia be given priority
assessment during clinical diagnosis of sleep disorders in stroke.
Keywords: insomnia, sleep disturbances, sleep quality, stroke.
2
RESUMO
As queixas sobre os problemas do sono não têm sido bem identificadas após um Acidente
Vascular Encefálico (AVE). O objetivo deste estudo foi investigar os fatores preditivos da
qualidade do sono e queixas de insônia em pacientes com AVE. Participaram 70 indivíduos,
40 pacientes (57 ± 7 anos) e 30 controles saudáveis (52 ± 6 anos) que foram avaliados pelo
Índice de Qualidade de Sono Pittsburgh (IQSP) e Questionário de Hábitos do Sono. Os dados
foram analisados pelo teste t de Student e regressão logística. Em média, os pacientes
apresentaram má qualidade do sono (pacientes: 6,3 ± 3,5; controles: 3,9 ± 2,2, p = 0,002) e a
queixa de insônia foi a mais prevalente (pacientes: 37,5%; controles: 6,7%, p = 0,007). A
ausência de queixa de insônia (OR = 0,120 IC 95% = 0,017-0,873, p = 0,036) e a diminuição
da latência do sono (OR = 0,120 IC 95% = 0,017-0,873, p = 0,036) foram os fatores de
proteção da qualidade do sono. Sexo feminino (OR = 11,098; IC 95% = 1,167-105,559, p =
0,036) e sono fragmentado (OR = 32,040; IC 95% = 3,236-317,261, p = 0,003) foram os
fatores de risco para a queixa de insônia. Nós sugerimos que as queixas de má qualidade do
sono e da insônia sejam priorizadas durante o diagnóstico clínico dos distúrbios do sono no
AVE.
Palavras-chave: insônia, distúrbios do sono, qualidade do sono, Acidente Vascular Encefálico.
3
1. INTRODUCTION
Stroke, defined as a focal disturbance of cerebral function as a result of vascular
alterations, is a pathology that represents a serious public health problem in Brazil, given that
the country has the highest death rate by stroke in Latin America and that it is still neglected
by the population and even by health professionals (Lotufo 2005; Pontes-Neto et al. 2008;
Coca et al. 2008).
In addition to physical and cognitive impairment, the occurrence of sleep disorders is
an important aspect that needs to be considered in the clinical approach to patients with
stroke. The rehabilitation process, for example, which occurs from the initial stage following
the stroke, is continuous and prolonged in many cases, and may be compromised if patients
experience poor sleep quality or sleep disturbances, mainly because sleep is a function that
interferes significantly in the learning and memory consolidation processes (Curcio et al.
2006; Wright Jr et al. 2006).
The diagnosis and treatment of sleep disorders is a field that has been expanding
considerably in recent years, with studies on the physiology and physiopathology of sleep.
The alterations in the amount, quality or regulation of sleep include: insomnia, hypersomnia,
obstructive sleep apnea syndrome, restless legs syndrome, circadian rhythm disorders, which
encompass, among others, jet lag, delayed and advanced sleep phase syndrome. On the other
hand, the alterations in behavioral or physiological events that occur during sleep in general,
in specific stages of this sleep or in sleep-wake transitions, these include: sleep walking, night
terrors, nightmares, sleep-talking, teeth grinding, nocturnal enuresis and snoring, among
others (AASM 2001).
Studies on patients with stroke show the occurrence of disturbed sleep, the main
disorder being obstructive sleep apnea syndrome, which occurs in 60% to 90% of patients
(Koch et al. 2007; Srijithesh et al. 2011). Complaints of insomnia and excessive daytime
4
somnolence are also found in this clinical population (Ferrea et al. 2010). Campos et al.
(2005) showed various episodes of dozing at different times and poor sleep quality in chronic
stage stroke patients.
Literature studies related to sleep in stroke patients have little mention of sleep
problem complaints and their relationship with sleep quality. We believe that the
identification of sleep disturbances in patients with stroke, by means of reports and
complaints, is crucial in the diagnostic and therapeutic clinical approach. Self-reporting
methods, such as sleep questionnaires, measure the sleep quality experienced by the patient,
considering both the quantitative and qualitative aspects of sleep are important to identify the
possible complaints of sleep disorders earlier. Moreover, these subjective methods are easily
managed, inexpensive and of wide applicability, in both clinical practice and research
(Chartier-Kastler and Davidson 2007; Gallasch and Gradisar 2007). The identification of
impaired sleep quality and frequency of sleep disturbance complaints in patients with stroke
may lead to corrective measures, such as the establishment of proper routines and habits, as
well as more appropriate times for therapeutic interventions. Thus, the aim of this study was
to investigate the predictive factors of subjective sleep quality and of insomnia complaints in
patients with stroke.
2. MATERIALS AND METHODS
2.1 Sample
The sample was composed of 70 subjects, including a group of 40 patients with
diagnosis of unilateral stroke enrolled at physical therapy services in the city of Natal, Brazil
who were compared to a group of 30 healthy employees of the Universidade Federal do Rio
Grande do Norte. The group of patients was composed of 27 men and 13 women, aged
between 45 and 65 years (57 ± 7 years), lesion time between 1 and 36 months (11 ± 9
5
months), with 24 and 16 patients presenting with impaired right and left cerebral hemisphere,
respectively. The group of healthy individuals was composed of 15 men and 15 women, aged
between 45 and 64 years (52 ± 6 years). The schooling level of the participants varied
between primary and secondary education. The exclusion criteria adopted for the patients
were: recurrent cerebral lesion, serious cognitive disorders, aphasia and the use of
tranquilizers, antidepressants or neuroleptics. For the controls, individuals with cognitive
disorders, shift workers or those who had taken a recent transmeridian trip were excluded.
2.2 Procedures
The study was approved by the institutional Research Ethics Committee. The
participants were informed of the study procedures and asked to sign a consent form. The
patients were interviewed and provided the following information: identification, results of
computerized tomography, history of the disease, risk factors present, medication used and
lesion time. The Cumulative Illness Rating Scale (CIRS), an assessment scale of
comorbidities, was used with the healthy individuals. This scale consists of a standardized
clinical evaluation that investigates the overall health data of the individual in six organic
systems, with scores ranging from 0 to 4, to ensure the current healthy status of the individual
(Parmelee et al. 1995). To determine the degree of neurological impairment of the patients,
we used the National Institute of Health Stroke Scale (NIHSS) protocol validated in Brazil by
Caneda et al. (2006). The scale is composed of 11 items that assess awareness level, eye
movements, visual field, facial movements, motor function and upper and lower limb ataxia,
as well as sensitivity, language, presence of disarthria and visual-spatial neglect. The total
score, ranging from 0 to 42 points, is the sum of the points of each item, in which higher
scores indicate greater neurological impairment.
6
The Horne-Östberg Questionnaire (Horne and Östberg 1976) was then applied to
identify the chronotype of the subjects. This questionnaire is composed of 19 questions about
preferable times for going to sleep and waking up. Finally, a subjective assessment of sleep
was conducted using the Pittsburg Sleep Quality Index (PSQL) (Buysse et al. 1989),
validated in Brazil by Ceolim (1999) and the Sleep Habits Questionnaire (Vinha et al. 1997).
The PSQL is a questionnaire used to assess the sleep quality of an individual through 7
components: subjective sleep quality, latency, duration, efficiency, sleep disorders, use of
sleep medication and daytime dysfunction. The scores of all the components are added to
obtain a value between 0 and 21, in which scores higher than 5, indicate poor sleep quality.
The Sleep Habits Questionnaire is a validated and standardized instrument with 47 questions
that aims to draw a sleep-wake profile of the subject assessed. In this study information from
this instrument about sleep was used to determine complaints of: insomnia (question 16),
difficulty falling asleep (“yes” answer on question 18), fragmented sleep (“yes” answer on
question 22), sleep apnea (“yes” answer on question 26), restless legs syndrome (“moving the
legs” answer on question 28), nightmares (“yes” answer on question 24), teeth grinding (
“grinding teeth” answer on question 28), sleep talking (“talk in sleep” answer question 28),
snoring (“snoring “ answer on question 28), sleep walking (“sleep walk” answer on question
28), rhythmic movement disorder (“hit head” answer on question 28) and night terrors
(“scream while sleeping” answer on question 28).
2.3 Data Analysis
Data analysis was conducted using SPSS 15.0 (Statistical Package for the Social
Science Software) and a significance level less than 5% for all the statistical tests. The chisquare test, with Yates’ correction (when necessary) was used to compare the patients and
controls in terms of the absolute and percentage frequency of the variables: sex, age,
7
schooling and chronotype, as well as, the frequency of sleep disorder complaints. After
normal distribution was confirmed (Kolmogorov-Smirnov test) in the PSQI scores and their
components, non-paired Student’s t-test was applied to compare the patients and the healthy
individuals. In the group of patients the chi-square test was used to determine the association
between sleep quality and the clinical variables and between insomnia complaint and the
demographic and clinical variables, PSQI categories, and other complaints of sleep disorder.
Logistic regression analysis was carried out using the stepwise backward method, to identify
the predictive factors of subjective sleep quality (with the categorized PSQI score) and of
insomnia complaint in the patients.
3. RESULTS
The chi-square test showed no significant difference between the groups in terms of the
absolute and percentage frequency for sex, age, schooling, and chronotype (Table 1).
However, the non-paired Student’s t-test did show a significant difference with respect to the
mean PSQI score, with a higher mean score for the patients (6.3  3.5) compared to the
controls (3.9  2.2) (p= 0.002). In regard to the PSQI scores, a significant difference was
found between patients and controls in terms of latency (p= 0.019), duration (p= 0.039) and
daytime dysfunction (p= 0.001). The main complaints reported by the participants were:
insomnia, apnea, restless legs, nightmares, teeth grinding, sleep talking and snoring, but a
significant difference was found only in the frequency of insomnia complaint in the patients
(37.5%) and controls (6.7%), with the patients showing greater frequency (p= 0.007). The
association between sleep quality and clinical variables was investigated and the result
obtained show no significant difference between the patients with good and poor quality with
respect to lesion time (p = 0.746), cerebral hemisphere affected (p= 0.601) and neurological
degree (p= 0.749).
8
Logistic regression analysis to identify the predictive factors of sleep quality in the
patients was conducted with the variables that were significant in the previously mentioned
Student’s t-test and chi-square test. Two models were built and the second better identified
the variables related to sleep quality. Significant values were found for insomnia (among the
sleep disorder complaints) and latency (among the PSQI components). The upper and lower
limits of the confidence interval for the Odds Ratio showed that absence of insomnia
complaint and decreased latency were the main protective factors of sleep quality (Table 2).
To select the variables associated with insomnia complaint in the patients, an analysis
was conducted using the chi-square test with the following variables: sex, age, schooling,
chronotype, lesion time, cerebral hemisphere affected and neurological degree. The analysis
shows that the female patients (53.3%; p= 0.029) and those with impaired left cerebral
hemisphere (60%; p= 0.045) had more insomnia complaints. A significant association was
observed between insomnia complaint in the patients and the PSQI components subjective
sleep quality (86.7%; p= 0.01) and sleep latency (66.7%; p= 0.007), showing that the higher
frequency of compromised subjective sleep quality and latency were in the group of patients
complaining of insomnia.
Analysis of the association between insomnia complaint in the patients and the other
sleep disorders showed a greater number of complaints regarding difficulty in falling asleep
(53.3%; p= 0.029) and of fragmented sleep (86.7%; p= 0.001) in the patients with insomnia
complaints than in those with no such complaints.
The logistic regression models for insomnia complaints in the patients were
constructed with variables that showed significant association, such as: sex, cerebral
hemisphere affected, subjective sleep quality, latency, difficulty in falling asleep, and
fragmented sleep. Of the five models built, the last best identified the variables related to
insomnia complaint. The lower and upper limits of the confidence interval for the Odds Ratio
9
(OR) indicated that the female sex and fragmented sleep can be considered risk factors for
insomnia complaint in the sample studied (Table 3).
4. DISCUSSION
The results obtained for the PSQI score suggest that the patients, on average,
experience poor sleep quality and had a higher latency, duration, and daytime dysfunction
compared with control subjects. These data corroborate the results of an earlier study
conducted by Campos et al. (2005) on poor sleep quality and the longer sleep duration of
patients in the late stage of stroke recovery. This finding suggests that longer sleep duration is
likely due to compensation for the poor sleep quality that these patients experience, indicating
the possibility of behavioral modification after the cerebral lesion. In a study on drowsiness
with elderly adults, Chasens et al. (2007) confirmed a relationship between drowsiness and
sleep quality, given that individuals with daytime drowsiness reported worse sleep quality.
Accordingly, sleep quality assessment has even greater clinical relevance in terms of its early
identification in patients with a neurological lesion, such as those with stroke.
The results obtained for the frequency of sleep disorder complaints showed a
significant difference between patients and controls with respect to insomnia complaint.
Bassetti (2005) found that between 20% and 40% of patients experienced sleep-wake
disorders. The most frequently observed sleep disorder complaints were: insomnia, excessive
daytime drowsiness/fatigue, and hypersomnia (increased need of sleep). This author suggests
that the cerebral lesion itself, often at the thalamic or brain stem level, may be a cause for
persistent sleep-wake disorders. An analysis of only sleep complaint in the present study
shows a relevant frequency of 37.5% in the patients who suffered a stroke, but in only 6.7%
of the controls. These values corroborate the data of an earlier study by Leppavuori et al.
(2002) about the frequency of insomnia in patients with stroke, but with a higher relevant
10
frequency (56.7%), possibly due to the more advanced age range (55 to 85 years) and to the
shorter lesion time (3 to 4 months).
The obstructive sleep apnea, restless legs, nightmares, teeth grinding, sleep talking and
snoring studied did not differ statistically in frequency between the patients and controls,
likely because it is seldom self-reported, and for this reason both groups rarely mentioned
these complaints. According to Masel et al. (2001), the prevalence of sleep apnea syndrome,
periodic limb movement disorder and hypersomnia in patients with cerebral lesion, may cause
excessive daytime drowsiness. This finding suggests that polysomnographic assessment
should be considered as part of the investigation of patients with cerebral lesion to
complement the subjective evaluation.
A relationship between sleep quality and sex was reported by Wells et al. (2004) in
their assessment of patients with obstructive sleep apnea, in which women experienced worse
sleep quality than did men. However, these results were not found in the present study, likely
owing to a number of differences in the present investigation in terms of age range and
pathology. With respect to the clinical variables, no significant difference was found between
the patients with poor and good sleep quality as to lesion time, cerebral hemisphere affected
and neurological degree.
Among the predictive factors of sleep quality assessed, insomnia complaint and
latency were the most influential. In this case, the lack of insomnia complaint and decreased
latency may be considered strong protective factors of sleep quality (Figure 1), suggesting
that when the sleep quality of patients is evaluated, insomnia complaints and sleep latency
must be identified first.
Our study also proposed to investigate the predictive factors of sleep complaint. The
results of the associations between sleep complaint in the patients with stroke and the
demographic and clinical variables and PSQI components, showed significant differences.
11
These findings indicate that the presence of insomnia complaint in patients with stroke was
associated to the female sex, the impaired left cerebral hemisphere, presence of fragmented
sleep complaints, difficulty in falling asleep, alterations in sleep latency (greater sleep
latency) and subjective sleep quality (worse subjective sleep quality).
The study conducted by Leppavuori et al. (2002) analysed insomnia in patients with
stroke and observed that those with insomnia were more frequently women and older than
patients without insomnia, suggesting a relationship between insomnia complaint in patients
with stroke and the female sex as well as advanced age. This sex-related finding confirms the
association between insomnia complaint in patients with stroke and the female sex observed
in the present study. However, the age association observed by the aforementioned authors
was not found in the present study, likely because the sample study of Leppavuori et al.
(2002) was composed of older individuals than the patients of our study. The association
between insomnia complaint and impaired left cerebral hemisphere corroborates the results of
Leppavuori et al. (2002), in which patients with stroke and insomnia complaint suffered more
frequently from hemispheric impairment (left hemispheric lesion). With respect to the
association between insomnia complaint and the other sleep disturbances, a relationship was
found only between insomnia and fragmented sleep complaint and difficulty in falling asleep.
Despite the associations found, multivariate analysis showed that the female sex and
fragmented sleep complaint were risk factors for insomnia complaint in patients with stroke
(Figure 1).
The study has a number of limitations that must be considered. The sample of patients
could have shown greater variability in neurological degree, thereby enabling a more
thorough analysis of the association between sleep quality and sleep disorder complaints,
according to the level of clinical impairment. Another limitation was the impossibility of
correlating the cerebral areas affected with the sleep disorders of the patients, because the
12
neuroimaging examinations were performed at different hospitals, precluding standardization
of the medical reports. Despite these limitations, the study identified that insomnia complaint
and decreased latency were the most important predictive factors of poor sleep quality in the
patients. In addition, the female sex and fragmented sleep complaint can be considered the
main predictive factors for insomnia complaint in the sample studied. Future studies may be
conducted to assess if the implementation of sleep disorder treatment would effectively
improve the rehabilitation potential of patients with stroke.
Acknowledgements
This research was supported by PROAP/Capes.
REFERENCES
AMERICAN
ACADEMY
OF
SLEEP
MEDICINE
(AASM).
2001.
International
classification of sleep disorders, revised: diagnostic and coding manual. Chicago.
BASSETTI C. 2005. Sleep and stroke. Semin Neurol 25: 19-32.
BUYSSE DJ, REYNOLDS CF, MONK TH, BERMAN SR AND KUPFER DJ. 1989. The
Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research.
Psychiatry Res 28: 193-213.
CAMPOS TF, DIÓGENES FP, FRANÇA FR, DANTAS RCS, ARAÚJO JF AND
MENEZES AAL. 2005. The sleep-wake cycle in the late stage of cerebral vascular accident
recovery. Biol Rhyt Res 36: 109-113.
CANEDA M, FERNANDES J, ALMEIDA A AND MUGNOL F. 2006. Reliability of
neurological assessment scales in patients with stroke. Arq de Neuropsiquiatr 64: 690-697.
CEOLIM MF. 1999. Padrões de atividade e de fragmentação do sono em pessoas idosas. Tese
de Doutorado, Universidade de São Paulo.
13
CHARTIER-KASTLER E AND DAVIDSON K. 2007. Evaluation of quality of life and
quality of sleep in clinical practice. Eur Urol Sup 6: 576-584.
CHASENS ER, SEREIKA SM, WEAVER TE AND UMLAUF MG. 2007. Daytime
sleepiness, exercise, and physical function in older adults. J Sleep Res 16: 60-65.
COCA A, MESSERLI FH, BENETOS A, ZHOU Q, CHAMPION A, COOPER-DEHOFF
RM AND PEPINE PCJ. 2008. Predicting stroke risk in hypertensive patients with coronary
artery disease: a report from the INVEST. Stroke 39: 343-348.
CURCIO G, FERRARA M AND GENNARO LD. 2006. Sleep loss, learning capacity and
academic performance. Sleep Med Rev 10: 323-337.
FERREA A, RIBÓB M, RODRÍGUEZ-LUNAB D, ROMEROA O, SAMPOLC G,
MOLINAB CA AND ÁLVAREZ-SABINB J. 2010. Los ictus y su relación con el sueño y los
trastornos del sueño. Neurol 10: 1-16.
GALLASCH J AND GRADISAR M. 2007. Relationships between sleep knowledge sleep
practice and sleep quality. Sleep Bio Rhyth 5: 63-73.
HORNE JA AND ÖSTBERG O. 1976. A self-assessment questionnaire to determine
morningness-eveningness in human circadian rhythms. Int J Chronobiol 4: 97-110.
KOCH S, ZUNIGA S, RABINSTEIN AA, ROMANO JG, NOLAN B, CHIRINOS J AND
FORTEZA A. 2007. Signs and symptoms of sleep apnea and acute stroke severity: is sleep
apnea neuroprotective? J Stroke Cerebrovasc Dis 16: 114-118.
LEPPAVUORI A, POHJASVAARA T, VATAJA R, KASTE M AND ERKINJUNTTI T.
2002. Insomnia in ischemic stroke patients. J Stroke Cerebrovasc Dis 14: 90-97.
LOTUFO PA. 2005. Stroke in Brazil: a neglected disease. São Paulo Med J 123: 3-4.
MASEL BE, SCHEIBEL RS, KIMBARK T AND KUNA ST. 2001. Excessive daytime
sleepiness in adults with brain injuries. Arch Phys Med Rehabil 82: 1526-1532.
14
PARMELEE PA, THURAS PD, KATZ IR AND LAWTON MP. 1995. Validation of the
Cumulative Illness Rating Scale in a geriatric residential population. J Am Geriatr Soc 43:
130-137.
PONTES-NETO OM, SILVA GS, FEITOSA MR, FIGUEIREDO NL, FIOROT JR JA,
ROCHA TN, MASSARO AR AND LEITE JP. 2008. Stroke awareness in Brazil: alarming
results in a community-based study. Stroke 39: 292-296.
SRIJITHESH PR, SHUKLA G, SRIVASTAV A, GOYAL V, SINGH S AND BEHARI M.
2011. Validity of the Berlin Questionnaire in identifying obstructive sleep apnea syndrome
when administered to the informants of stroke patients. J Clin Neurosci 18: 340–343
VINHA D, CAVALCANTE J AND ANDRADE MMM. 2002. Sleep-wake patterns of
workers and non-workers students. Biol Rhythm Res 33: 417-426.
WELLS RD, DAY RC, CARNEY RM, FREEDLAND KE AND DUNTLEY SP. 2004.
Depression predicts self-reported sleep quality in patients with obstructive sleep apnea.
Psychosom Med 66: 692-697.
WRIGHT JR KP, HULL JT, HUGHES RJ, RONDA JM AND CZEISLER CA. 2006. Sleep
and wakefulness out of phase with internal biological time impairs learning in humans. J
Cogn Neurosci 18: 508-521.
15
Table 1. Comparison of absolute and percentage frequency of the variables sex, age,
schooling and chronotype, between patients and healthy controls, using the Chi-square test.
VARIABLES
Sex
Age
Schooling
Chronotype
PATIENTS
CONTROLS
N
%
n
%
p value
Female
13
32.5
15
50.0
0.139
Male
27
67.5
15
50.0
 60 years
15
37.5
5
16.7
 60 years
25
62.5
25
83.3
Primary
32
80.0
18
60.0
Secondary
8
20.0
12
40.0
Indifferent
4
10.0
4
13.4
Moderate Morning
18
45.0
13
43.3
Extreme Morning
18
45.0
13
43.3
0.060
0.067
0.910
16
Table 2. Logistic regression models (Odds Ratio, upper and lower limits of the 95%
confidence interval and p-value) considering insomnia complaint, subjective quality, latency,
duration and daytime dysfunction as independent variables and sleep quality as the dependent
variable.
VARIABLES
ODDS RATIO
LOWER
UPPER
p
Insomnia
0.118
0.016
0.895
0.036
Subjective quality
0.388
0.019
7.975
0.539
Latency
0.141
0.019
1.069
0.058
Duration
-
-
-
0.999
0.080
0.006
1.071
0.056
Insomnia
0.120
0.017
0.873
0.036
Latency
0.120
0.017
0.873
0.036
Duration
-
-
-
0.999
0.082
0.006
1.052
0.055
Model 1
Daytime dysfunction
Model 2
Daytime dysfunction
17
Table 3. Logistic regression models (Odds Ratio, upper and lower limits of the 95%
confidence interval and p-value) considering sex, cerebral hemisphere affected, subjective
sleep quality, latency, difficulty in falling asleep and fragmented sleep as independent
variables sand insomnia complaint as the dependent variable.
VARIABLES
ODDS RATIO
LOWER
UPPER
p
Sex
22.401
1.253
400.501
0.035
Cerebral hemisphere affected
3.208
0.380
27.108
0.284
Subjective quality
11.353
0.728
176.993
0.083
Latency
0.800
0.081
7.928
0.849
Difficulty in falling asleep
3.835
0.458
32.119
0.215
Fragmented sleep
22.789
1.399
371.300
0.028
Sex
8.567
0.745
98.483
0.085
Cerebral hemisphere affected
2.338
0.334
16.371
0.392
Lantency
1.567
0.195
12.587
0.672
Difficulty in falling asleep
3.028
0.454
20.194
0.252
Fragmented sleep
20.776
1.729
246.725
0.017
Sex
9.991
0.967
103.213
0.053
Cerebral hemisphere affected
2.158
0.327
14.264
0.425
Difficulty in falling asleep
3.392
0.547
21.037
0.190
Fragmented sleep
25.862
2.467
271.126
0.007
Sex
11.578
1.171
114.518
0.036
Difficulty in falling asleep
3.683
0.609
22.290
0.156
Fragmented sleep
29.733
2.763
319.919
0.005
Sex
11.098
1.167
105.559
0.036
Fragmented sleep
32.040
3.236
317.261
0.003
Model 1
Model 2
Model 3
Model 4
Model 5
18
STROKE
SLEEP
QUALITY
INSOMNIA
Protective factors
Risk factors
 Sleep Latency
Female sex
Absence of insomnia complaint
Fragmented Sleep
Figure 1- Schematic drawing of the predictors of quality of sleep and insomnia complaints of
patients with stroke. On average, the patients showed poor sleep quality and insomnia
complaint was the most prevalent. The decreased latency of sleep and the absence of insomnia
complaint were the protective factors of sleep quality. Female sex and fragmented sleep were
the risk factors for insomnia complaint.
19